Although our methods for measuring quality of care in nursing homes have improved substantially, we still lack a clear understanding of how nursing home structure and local community context may influence quality of care. We also lack an understanding of how segregation of access to nursing homes by minority elders is related to differences in quality of care received by white and minority nursing home residents. The specific aims of this study are: 1) To quantify the extent to which access to nursing home care is segregated by race/ethnicity; 2) Using state-of-the-art indicators of care quality in long term care nursing facilities, develop a measure of racial/ethnic disparities in care quality that compares the odds of receiving poor quality care for whites and nonwhites; 3) To examine variation in and relationships between the two separate concepts of nursing home quality of care and racial/ethnic disparities in nursing home quality of care; 4) Using a smaller subset of facilities from this population that are participating in an existing panel study, to test for the effects of demographic "mismatch" between community and nursing home staff on racial/ethnic disparity in nursing home care quality; and 5) To develop and test a multilevel model explaining variation in nursing home quality of care as well as racial/ethnic disparities in nursing home quality of care as a function of patient level, facility level and community level factors. We propose to use recently validated measures of care quality to develop racial/ethnic disparity measures in 6 different areas of care (pain management, pressure ulcers, depression management, use of restraints, use of antipsychotic drugs and incontinence management), and we will compare the quality of nursing home care received by whites, African Americans, Native Americans and non-white Hispanics. The Strategic Assessment of Geriatric Drug Use and Epidemiology (SAGE) Database will provide the core data for this project on over 7000 nursing homes in seven different states, with a smaller subanalysis of 308 nursing homes from the Do Good Nursing Homes project.